Detection of COVID-19 and its pulmonary stage using Bayesian hyperparameter optimization and deep feature selection methods

Since the first case of COVID-19 was reported in December 2019, many studies have been carried out on artificial intelligence for the rapid diagnosis of the disease to support health services. Therefore, in this study, we present a powerful approach to detect COVID-19 and COVID-19 findings from computed tomography images using pre-trained models using two different datasets. COVID-19, influenza A (H1N1) pneumonia, bacterial pneumonia and healthy lung image classes were used in the first dataset. Consolidation, crazy-paving pattern, ground-glass opacity, ground-glass opacity and consolidation, ground-glass opacity and nodule classes were used in the second dataset. The study consists of four steps. In the first two steps, distinctive features were extracted from the final layers of the pre-trained ShuffleNet, GoogLeNet and MobileNetV2 models trained with the datasets. In the next steps, the most relevant features were selected from the models using the Sine–Cosine optimization algorithm. Then, the hyperparameters of the Support Vector Machines were optimized with the Bayesian optimization algorithm and used to reclassify the feature subset that achieved the highest accuracy in the third step. The overall accuracy obtained for the first and second datasets is 99.46% and 99.82%, respectively. Finally, the performance of the results visualized with Occlusion Sensitivity Maps was compared with Gradient-weighted class activation mapping. The approach proposed in this paper outperformed other methods in detecting COVID-19 from multiclass viral pneumonia. Moreover, detecting the stages of COVID-19 in the lungs was an innovative and successful approach.     

Intrinsic plagiarism analysis

Research in automatic text plagiarism detection focuses on algorithms that compare suspicious documents against a collection of reference documents. Recent approaches perform well in identifying copied or modified foreign sections, but they assume a closed world where a reference collection is given. This article investigates the question whether plagiarism can be detected by a computer program if no reference can be provided, e.g., if the foreign sections stem from a book that is not available in digital form. We call this problem class intrinsic plagiarism analysis; it is closely related to the problem of authorship verification. Our contributions are threefold. (1) We organize the algorithmic building blocks for intrinsic plagiarism analysis and authorship verification and survey the state of the art. (2) We show how the meta learning approach of Koppel and Schler, termed “unmasking”, can be employed to post-process unreliable stylometric analysis results. (3) We operationalize and evaluate an analysis chain that combines document chunking, style model computation, one-class classification, and meta learning.     

Selective separation of cobalt and nickel by flat sheet supported liquid membrane using Alamine 300 as carrier

Cobalt and nickel are among the most important nonferrous metals. The using of flat sheet supported liquid membranes (FSSLMs) to remove metals from wastewaters has been used actively by the scientific and industrial communities. In this study, the selective separation of cobalt from thiocyanate solutions containing cobalt and nickel by FSSLM was examined using tri-n-octylamine (Alamine 300) as carrier. The FSSLM was consisted of extractant, flat sheet support and organic solvent. The various parameters were studied to determine the optimum extraction and striping conditions of cobalt and nickel. These parameters were stirring speeds of phases, NH₄SCN concentration, pH, diluent type, extractant concentration, stripping reagent concentration and modifier concentration. Concentration of cobalt and nickel were determined by Shimadzu AA-6701GF spectrophotometer. In the optimum conditions, selective separation of cobalt was achieved with an efficiency of 98.4% within 8 h, for equimolar feed mixtures, 400 mg/L Co + 400 mg/L Ni, and the separation factor of Co(II) over Ni(II) was 234.4. In addition, for nonequimolar feed mixtures, 500 mg/L Co + 1000 mg/L Ni, Ni in excess, selective separation of cobalt was 99.9%, and the separation factor of Co was 506 in the same time.     

Selective separation of cobalt and nickel by supported liquid membranes

The selective separation of cobalt from acidic media, containing both equimolar and nonequimolar mixtures of cobalt and nickel, was examined by supported liquid membranes using Alamine 336 as mobile carrier dissolved in various diluents. The membrane support was microporous hydrophobic polypropylene Celgard 2500 (25 μm thick, 0.209 × 0.054 μm pore size and 55% porosity). Acetic acid-Na acetate buffer was used for the adjustment of the feed pH which was critical. Various parameters were experimentally studied and the optimum conditions were determined.     

Characterization and elemental analysis of wood pellets obtained from low-valued types of wood

In this study, we compared the quality of wood pellets obtained from several different raw materials, i.e., Rhododendron ponticum (Type 1), Laurus nobilis (Type 2), and Castanea sativa (Type 3). The quality of the wood pellets was characterized mainly by their bulk density, moisture content, ash content, volatiles, sulfur content in the ash, total sulfur content, heating values, elemental analysis of the ash, and chlorine content. The results showed that bulk density was similar for each type of pellet. In quality values, ash content and the sulfur content in the ash were found to be lower for Type 3 (Chestnut wood pellets) than they were for the other two types. The results also showed that dry samples of Type 1, Type 2, and Type 3 wood pellets had heating values of 5057, 4691, and 4571 kcal/kg, respectively, whereas the original (undried) samples had heating values of 4571, 4409, and 4293 kcal/kg, respectively.     

Physical and chemical properties of chufa (Cyperus esculentus L) tubers grown in the Çukurova region of Turkey

Four lines of chufa (Cyperus esculentus) grown in 1998 and 1999 in the Çukurova region of Turkey were analysed for their physical properties, proximate chemical composition and fatty acid contents. The chufa lines contained on average (g kg^?1) 932.8 dry matter, 245.0 crude lipid, 256.8 starch, 14.3 ash, 50.5 protein, 89.1 crude fibre, 17.1 reducing sugar, 154.3 total sugar and 130.4 sucrose. Hunter L, a⁺ and b⁺ colour values of ground chufa samples were in the ranges 55.93–60.59, 3.71–5.09 and 15.60–16.85 respectively. Individual chufa tubers weighed between 0.224 and 0.283 g. The fatty acid composition of chufa oil included (g kg^?1) 689.2–732.9 oleic acid, 125.5–141.2 palmitic acid and 99.6–154.6 linoleic acid, which is comparable with that of olive oil. After storage for 1 year the differences in mean values were significant (p < 0.05). © 2002 Society of Chemical Industry     

The extraction of lactic acid by emulsion type of liquid membranes using alamine 336 in escaid 100

The extraction of lactic acid from aqueous solutions through an emulsion liquid membrane containing Alamine 336 as carrier was investigated. The influence of mixing speed, diluent type, surfactant concentration, extractant concentration, feed solution pH, stripping concentration, phase ratio, and feed concentration were examined. Liquid membrane consists of a diluent (n‐heptane, toluene, kerosene, Escaid 100, and Escaid 200), a surfactant (Span 80) and an extractant (Alamine 336), and Na₂CO₃ were used as a stripping solution. It is possible to extract 91% of lactic acid from aqueous solutions using Alamine 336 in Escaid 100, as an extractant and a diluent respectively.